IT MUST be rare for a small paperback book aimed at the popular market to introduce a radical new idea into medical thought. Yet most of the British doctors now studying and treating food intolerance (see last week’s Â鶹´«Ã½) trace their interest back to the publication of Not All In The Mind by Richard Mackarness, in 1976. Doctors in the US had first recognised food intolerance more than 50 years earlier, but it was Mackarness, a psychiatrist and physician at Basingstoke District Hospital, who was responsible for introducing the concept to Britain.
In Not All In The Mind, Mackarness tells the following tale about one of his former colleagues in the psychiatric department. During a lecture by Mackarness on food intolerance – also known as ‘masked food allergy’ – the psychiatrist began to wonder if his severe and unexplained fatigue might be due to some such cause. According to Mackarness, fatigue could be a symptom of food intolerance, and the sufferer might crave the particular food causing the problem. The only food for which the psychiatrist had any craving was bacon, and although sceptical, he decided to try cutting out bacon, and other forms of pork, for a while. The result took him completely by surprise. Initially he felt a great deal worse, but within a week his fatigue had completely disappeared and his former energy returned. When he told his colleagues about his ‘cure’, they reacted, not surprisingly, with total disbelief. Convinced that the improvement in his health must be due to some other cause, they arranged for the hospital cook to adulterate the steak pie she was preparing for lunch with some finely chopped bacon. The subject of the experiment tucked into the pie unsuspectingly, under the watchful eyes of his colleagues. To their collective astonishment, halfway through the meal he pushed his plate aside, laid his head down on the table and fell soundly asleep. In Mackarness’s words: ‘The doctors who were in on the experiment were impressed and some of them even admitted that there might be something to masked food allergy after all.’
Advertisement
A decade later, the idea of food intolerance has much wider acceptance among doctors, but the orthodox medical view is still one of scepticism, despite several impressive double-blind controlled studies. Much of the medical establishment’s reluctance to accept food intolerance stems from the lack of any obvious mechanism: how on earth can bacon make a person fall asleep? How, for that matter, can common foods such as wheat, milk or eggs cause migraine, headaches, joint pains, diarrhoea, hyperactivity, fatigue, asthma, eczema or a constant runny nose? Moreover, why should a different cluster of symptoms appear in every patient? Those doctors who study food intolerance, and who are convinced that it is a real phenomenon, realise that they must demonstrate the underlying mechanisms responsible for such complex reactions if they are to convince their colleagues.
The study of food intolerance is historically linked with that of food allergy, and until relatively recently most attempts to find a mechanism centred on the immune system. The cause of classical or true allergies, including those to food, is the production of IgE antibodies to the offending molecule, or antigen . IgE binds to mast cells, and if its molecules then become cross-linked by binding to their particular antigen (a food protein, for example), this causes the mast cell to release various mediators, or chemical messengers. These mediators include histamine, serotonin and prostaglandins, and they have a dramatic effect on the body. In small quantities, they cause intense local inflammation. In larger amounts, they enter the circulation, making the capillaries dilate and become more permeable, so that there is a sudden drop in blood pressure. This reaction, known as anaphylactic shock, can be fatal.
In individuals with true food allergy, laboratory tests show high levels of total IgE, as well as IgE to specific foods. But this is not the case in food intolerance: in fact, the involvement of IgE is the crucial distinction between the two. Although some experiments have shown slight increases in IgE for the culprit foods in patients with migraine, it seems unlikely that IgE is the predominant cause of most food intolerance.
Where IgE may be important, however, is in the wall of the gut. The drug sodium cromoglycate is known to prevent mast cells from releasing their mediators (degranulating). Once food-intolerant patients have identified the foods that trigger their symptoms and eliminated them from the diet, they remain sensitive to those foods for a period of several months. Eating the food will provoke the symptoms, but if the patient takes sodium cromoglycate shortly before consuming the food, there is no response. Jonathan Brostoff, of the Middlesex Hospital in London, has tested sodium cromoglycate on his food-intolerant patients, and believes that a reaction with IgE and mast cells in the gut wall may underlie some food intolerance. When the patient eats the offending food, it could combine with IgE antibody in the gut walls and thus set off localised inflammation. The amount of IgE present is presumably fairly small, so that the acute reactions typical of true food allergy do not occur. But the inflammation could be sufficient to make the gut wall more permeable, allowing more indigested food molecules to get through. Once absorbed, those molecules might cause symptoms in a variety of ways.
Several studies substantiate this idea by showing that patients with food intolerance generally have more leaky guts than healthy individuals, although it is certainly not true of all patients. Indeed, those with symptoms affecting the gut alone often show fewer intact food molecules in the blood than normal people. In these cases, the molecules could be locked in immune complexes with antibodies in the gut wall, causing only localised inflammation.
Where intact food molecules do enter the blood, they usually form circulating immune complexes (CICs) with antibodies. Such complexes are not an unhealthy sign: everyone has some after a meal, because even healthy individuals absorb a few incompletely digested molecules. The formation of immune complexes is a necessary prelude to the removal of these molecules from the bloodstream by phagocytic cells: such cells engulf CICs and destroy them. But food-intolerant individuals tend to have larger CICs than other people, and more of them. Furthermore, the antibodies involved are not exactly the same as in normal individuals. Although researchers disagree about the details, it seems that in the food-molecule CICs of normal individuals, IgG and IgA predominate, whereas in food-intolerant individuals there is less IgA, more IgG and some IgE.
These differences could be important, because they make the CICs more likely to promote inflammation. IgE is present, and could trigger mast cells, whereas IgA, which is largely non-inflammatory, is under-represented.
But it is by no means certain that these abnormal CICs actually do any damage in the patient with food intolerance. To do harm, they would have to form deposits in the blood vessels, causing inflammation of the blood-vessel wall. This sort of reaction is a feature of more serious diseases, such as the autoimmune disorder, systemic lupus erythematosus (SLE). Patients with SLE form antibodies to their own proteins and have so many immune complexes in their blood that the phagocytic cells are unable to cope. Deposits form wherever there is turbulent flow through tiny blood vessels, such as in the kidneys, the joints, the skin and around the lungs. Rashes, joint pains and fever are among the symptoms.
Some doctors see a parallel with food intolerance, and have suggested that a similar reaction, but in a very much milder form, could be occurring, except that the antibodies are binding to food molecules rather than to the body’s own molecules. This might account for the joint pains seen in some patients, and, perhaps, for migraine, since this involves blood vessels in the brain. But at present there is no concrete evidence to support this idea.
IgE and CICs are not the only means by which the immune system could generate some of the symptoms of food intolerance. Another possible mechanism is via lymphokines or cytokines, hormone-like peptides that form a communications link between different cells of the immune system. Doctors first noticed their adverse effects when one type of lymphokine, interferon, came into use as a drug for hepatitis B. The side effects of interferon are legion and range from headache, lethargy and dizziness, to abdominal discomfort, bowel disturbance, nausea and joint pain. Doctors speculate that the release of lymphokines may cause much of the malaise of influenza and other infections, rather than the toxins that the pathogen itself produces. Lymphokines might also be responsible for the mysterious ailment called postviral syndrome or mylagic encephalitis. In the case of food intolerance, an abnormal immune response to food molecules could lead to the release of lymphokines and thus to the unpleasant symptoms. Although there is little direct evidence for this, several research groups around the world are now investigating the idea.
While the immune system undoubtedly plays a part in some cases of food intolerance – and perhaps in most cases – few doctors involved in this field now believe that it is the whole story. Some symptoms are difficult to reconcile with immunological causes, particularly the phenomenon of addictive eating: about half of all food-intolerant patients crave the food or foods that cause their symptoms. In many cases this craving is for wheat or milk and may go unnoticed because these foods tend to appear in every meal. But a few patients have rather odd cravings: some need eggs at every meal, and one patient, seen by Ronald Finn of the Royal Liverpool Hospital, ate raw potato daily.
A possible explanation for this bizarre feature of food intolerance has recently emerged. The incomplete digestion of food proteins produces certain peptides that can mimic the body’s own peptide hormones. Laboratory investigations show that some of these peptides can bind to the brain’s receptors for endogenous opioids, or endorphins, and they are therefore known as exorphins. Partial digests of wheat, maize and milk contain these exorphins, and other foods may also produce them, although a more extensive search among foodstuffs has yet to take place. It is not yet certain that exorphins do actually reach the right receptors in the body, but one piece of circumstantial evidence suggests that they do. Researchers have shown that partially digested wheat increases the transit time of food in the gut, and naloxone, a drug that binds to opioid receptors, can block the effect.
So it seems that while a fix of wheat and milk is unlikely to get you high – the amounts of exorphin involved are much too small – these foods might induce a sense of wellbeing in susceptible individuals. But the conclusive proof that they act as opioids in the brain and produce addictive eating, or any other symptoms, is still lacking. A shortage of funds for research into food intolerance (an unpromising area for pharmaceuticals companies, as treatment does not generally involve drugs) means that progress in investigating these novel ideas is bound to be slow.
If food can yield opioids and other hormone-like peptides, why do they affect some people but not others? No one knows the answer to this, but the greater permeability of the gut in those with food intolerance may be one factor – more peptides get though the gut wall and into the bloodstream. Another possibility is that enzyme deficiencies are partially to blame. If certain digestive enzymes are in short supply in the gut or operate inefficiently, this might lead to more peptides being available for absorption. Deficient enzymes could also allow the peptides to survive longer in the bloodstream.
Although the role of inadequate enzymes in the exorphin story remains speculative, there is little doubt that enzyme deficiencies are relevant to food intolerance as a whole. No one has yet looked in detail at digestive enzymes in the gut, but there is a growing body of evidence concerning enzymes that detoxify compounds absorbed into the bloodstream.
Research into the enzyme capabilities of patients with food intolerance has so far identifed at least three types of enzyme deficiency. The first concerns phenol sulphotransferase (PST) which is responsible for the first step in the degradation of phenols. People with food-induced migraine tend to be deficient in one type of PST, called PST-P. No one knows how phenols might cause migraine, but they could have a toxic effect if present in large quantities, or they could act as haptens, molecules that are too small to be antigens on their own, but which can act as antigens when combined with proteins.
Phenols may also be important in hyperactivity in children, because of high levels of a phenolic compound, p-cresol, appear in the faeces of such children. Bacteria in the gut produce this compound, but PST-P should break it down. The suspicion is that hyperactive children are defective in this enzyme, just as migraine patients are. Celia Gibb of Queen Charlotte’s Hospital in London, who is studying this possibility, has also found that some artificial food colourings are potent inhibitors of PST-P in vitro. If they can also inhibit PST-P in vivo, this might explain the putative link between hyperactivity and highly coloured ‘junk foods’: children with a defective form of PST-P would be made much, worse by chemicals that inhabit the emzyme’s activity.
A second type of enzyme deficiency concerns the monoamine oxidases, which remove the amine group of certain compounds such as tyramine and phenylethylamine. Migraine patients also tend to be defective in these enzymes. Unlike phenols, these amines have a well-understood mode of action: they affect the blood vessels in the brain, causing them to contract. A ‘rebound reaction’ then sets in, during which the blood vessels dilate excessively, causing the pain and visual disturbances of migraine. Foods that are notorious as triggers of migraine – cheese, citrus fruits, chocolate and red wine – are particularly rich in vasoactive amines. However, people who are affected by these foods are no more likely to have monoamine oxidase defects than others who suffer from migraine, so this is obviously not the whole story.
The third type of enzyme deficiency concerns the oxidation of sulphide groups to sulphoxide in a variety of compounds. A series of enzymes catalyse this reaction and doctors measure their efficiency by administering a harmless test drug, carboxymethylcysteine, and measuring the ratio of sulphoxide to sulphide excreted in the urine. Among the population as a whole, about 20 per cent of people perform poorly in this test, whereas in those with food intolerance the figure is 80 per cent. Taking only those patients who show sensitivity to food and to common synthetic chemicals (such as chlorine, natural gas, exhaust fumes and solvents), the incidence of poor sulphoxidation rises to 90 per cent.
This finding is of particular interest because the association between food intolerance and chemical sensitivity has always been difficult to explain in immunological terms. Theron Randolph, the American pioneer of research into food intolerance, first noted chemical sensitivity in some of his patients in 1951, and many other doctors have subsequently confirmed his findings. The people worst affected are often severely ill, and some well-publicised cases have inspired sensational newspaper stories about ‘total allergy syndrome’ and ‘allergies to the 20th century’.
Enzyme deficiencies appear to offer an explanation for the observed link between food intolerance and chemical sensitivity. The detoxification of synthetic compounds probably depends on enzymes whose original function was to break down bacterial products and certain substances in food – particularly in plant foods, because plants have a well-developed chemical armoury. Individuals who are less efficient metabolically than others have probably always existed, and natural selection would have weeded them out only or if they were severely affected. Those with slightly sub-optimal enzymes may have noticed few ill-effects until synthetic chemicals were introduced wholesale into the human environment. With this increased load, their enzyme deficiencies have become apparent, and their ability to deal with natural toxins in food may also have suffered, due to extraneous chemicals making impossible demands upon their detoxification systems. The case histories of many patients seem to support this idea: if they can avoid the chemicals to which they are sensitive, they become better able to cope with foods that were previously a problem.
Of the range of possible mechanisms to emerge from different research programmes – gut inflammation, CIC deposition, lymphokines, exorphins and enzyme deficiencies – none is likely to be the exclusive cause of food intolerance. Most doctors involved in this field believe that these diverse mechanisms could coexist and interact in a variety of ways, with one abnormality aggravating another. Indeed, it is possible that at least two defects, plus certain environmental triggers, are necessary for the symptoms of food intolerance to appear. This multifactorial explanation is attractive because it could resolve one of the paradoxes of food intolerance; the fact that no two patients have exactly the same symptoms would make sense if each has a different mixture of underlying causes.
Yet another possible mechanism is now emerging to add to this already complex picture. John Hunter of Addenbrooke’s Hospital in Cambridge has observed that many of his patients with irritable bowel syndrome (IBS) date the onset of their problem to a bout of gastroenteritis or a course of antibiotics. Investigating further, he has discovered that IBS is particularly likely to develop following hysterectomy. For this operation, many surgeons prescribe large doses of antibiotics as a preventive measure, before surgery takes place. A double-blind controlled trial with hysterectomy patients confirmed Hunter’s suspicions: those who did not take antibiotics before the operation rarely developed IBS.
Hunter believes that disturbances in the gut flora – the community of normally harmless commensal bacteria that inhabit our intestines – can explain these observations. Both antibiotics and the acute diarrhoea associated with gastroenteritis can depopulate the gut. Following such an upheaval the flora re-establishes itself, but it may have changed in composition, with more of one species and less of another, or with a new species gaining a foothold. Preliminary research indicates that the bacterial inhabitants of IBS sufferers are indeed different from those of healthy individuals.
How the disturbed flora might produce the symptoms of food intolerance is another matter. Hunter has suggested that a particular bacterial species – one that feeds on certain foodstuffs only – might become over-represented in the gut flora. The production of toxins, hormone-like peptides and other substances by the offending bacterium could then generate the symptoms, but only if the patient eats a particular food or foods – the food that the bacterium in question needs. At present, too little is known about the metabolism of the gut flora to decide if this is a likely scenario, but a research programme is now under way at Addenbrooke’s to test the hypothesis. Hunter, like his fellow researchers into food intolerance, does not rule out other mechanisms. Indeed, he believes that enzyme deficiencies could well play a part in the problem, with food-intolerant patients lacking enzymes that normally break down toxic bacterial products.
One organism that will come under scrutiny in Hunter’s research programme is a yeast, Candida albicans, that has generated a great deal of controversy among doctors treating food intolerance. Some, including Hunter, believe it is of little relevance to the problem. Others are convinced that it is a major factor, and perhaps the root cause of all food intolerance. According to the latter group, Candida can overpopulate the gut, especially if it receives a diet rich in sugars, and courses of antibiotics that knock out the bacterial population. Some believe that the contraceptive pill also favours its growth. Under favourable conditions, the yeast is said to convert from its normal single-celled form to a thread-like hyphal form that invades the gut wall and produces tiny holes, or microperforations. The theory is that these make the gut far more permeable, and that Candida toxins exacerbate the problem by generally depressing the immune system.
The available evidence suggests strongly that Candida can sometimes get out of control, and it probably causes serious symptoms in its own right, including diarrhoea, bloating, abdominal pain, psoriasis and depression. But the question of whether Candida causes microperforations in the gut wall and is an underlying cause of most food intolerance, as some practitioners claim, remains controversial.
Of all the unanswered questions about food intolerance, perhaps the most difficult concerns its prevalence, past and present. It is impossible to assess its incidence today – estimates range from an extremely cautious 1 per cent to a rather improbable 90 per cent, with most specialists settling for between 10 per cent and 30 per cent. The chance of estimating its incidence in earlier times is absolutely nil. Nevertheless, most of the doctors treating food intolerance and allergy have the impression that the incidence is increasing. And epidemiological studies from Africa show that one disorder positively linked with food intolerance, Crohn’s disease, is a feature of modern urban living.
Given the possible mechanisms of food intolerance, the idea that it is an epidemic of the late 20th century is plausible enough. Antibiotics, used widely only since the 1940s, can disturb the gut flora and, perhaps, lead to infestation by Candida. Some synthetic food colourings appear to inhibit an important enzyme, without which we cannot detoxify certain food compounds. An increased need to break down other synthetic chemicals, in our food, water and air, may contribute to the problems of individuals with enzyme deficiencies. Finally, the bottle-feeding of babies, combined with early weaning, has undoubtedly left some adults with a legacy of overreactive immune responses to food. Food intolerance could simply be a sign that the human body is not fully adapted to the novel environment we have created for ourselves.
Linda Gamlin is a freelance science writer and editor based in Bath, and author, with Jonathan Brostoff, of The Complete Guide to Food Allergy and Intolerance (Bloomsbury, Pounds sterling 9.95).
* * *
WHO’S WHO AMONG THE IMMUNOGLOBULINS
IMMUNOGLOBULINS are Y-shaped proteins found in the blood, gut lining, airways and other parts of the body. Specialised cells, called B lymphocytes or B cells, produce these proteins, which help to combat disease by combining with invading cells, viruses and toxins. They show a high degree of specificity for their target, the antigen.
Immunoglobulins, or Igs, are almost infinitely variable, thanks to a set of alternative genes for the antigen-binding region of the molecule. As the B cell matures, some unknown process randomly selects and combines these genetic building blocks. So each B cell produces its own unique Ig, which it carries on its surface.
If that Ig happens to bind to the proteins on an invading pathogen, or to some other extraneous molecule, the cell producing it begins to multiply and its daughter cells pump out free immunoglobulin, or antibody. These antibodies can combine with their antigen to form immune complexes. A complex set of control mechanisms govern this cell proliferation, and in the healthy person they prevent damaging reactions to benign antigens.
There are five major types of Ig, known as isotypes. The three that are of interest in food intolerance are IgG, IgE and IgA.
As antibodies, these different types of Ig have particular functions and properties. IgG is the most common, particularly in the blood. It consists of four subclasses, two of which (IgG1 and IgG3) can activate the complement system when they form immune complexes. The complement system is a highly complex sequence of reactions involving enzymes that attack and destroy the membranes of invading microorganisms. In the process, the host membranes also suffer, producing the visible signs of inflammation. During inflammation other cells of the immune system flood into the area to mount a concerted attack on potential pathogens.
IgE is much less common than IgG and probably has the specialised role of combatting multicellular parasites. Like other Igs, it emanates from B cells but later binds by its stem to the surface of mast cells and basophils.
This leaves the two branches of the molecule free to combine with antigen. When they do so, the cross-linking of the IgE molecules causes the cell to release powerful mediators, such as histamine, which initiate inflammation. The formation of IgE to intrinsically harmless antigens, such as pollen, house-dust mite or food molecules, is the cause of true or ‘classical’ allergies such as hay fever.
IgA occurs in the blood in small amounts, but is the predominant Ig of body secretions such as tears, saliva, milk, vaginal fluid and gastric mucus. Here it exists as two linked IgA molecules and is known as secretory IgA (SIgA). IgA has the special property of not activating the complement system when it binds to antigen, so that it does not cause inflammation.
* * *
NOT FIGHTING FOOD: THE ORAL TOLERANCE CONCEPT
RESEARCH into the causes of food intolerance has stimulated interest in a rather obvious but seldom-asked question: how does a healthy person’s immune system cope with the onslaught of millions of different food molecules? Early experiments in immunology showed that it was not simply a case of the gut wall rigidly excluding food antigens from the blood: the body actually learns not to react to food. One important series of experiments was carried out by H. G. Wells before he turned to writing as a career. Wells found that if he injected a food antigen such as egg white into a guinea pig or other experimental animal, it became sensitised to that antigen: a second injection would produce a massive immune reaction that was usually fatal (anaphylactic shock). However, if the guinea pig ate egg white before the first injection, it did not become sensitised and survived the second injection.
This phenomenon acquired the name of ‘oral tolerance’, but it is only recently that immunologists have begun to understand how it occurs. The induction of oral tolerance takes place in discrete areas in the gut wall, known as Peyer’s patches. Cells within each patch actively take up fluid from the gut in a process known as antigen sampling. Any antigen encountered provokes two separate responses, the first being the promotion of secretory IgA (SIgA) specific for that antigen. B-cells producing the appropriate SIgA proliferate and enter the lymphatic system which carries them to other secretory tissues – in distant parts of the gut, in the mammary glands, bronchi, lachrymal glands and elsewhere. By this means, all the body’s exposed surfaces (and the infant’s gut in the case of the breast-feeding mother) receive the protection of plentiful SIgA for that antigen. By binding to the antigen, SIgA greatly reduces its absorption through the gut wall, but without causing any inflammation.
The second response damps down the immune response to any molecules that do reach the blood. It involves two types of T cell that act as accessories in reactions between antigens and B cells (the ones that produce antibodies). One, known as a helper T cell, or Th cell, is essential for any B cell response. The other, known as a suppressor T cell, or Ts cell, can prevent a B cell from responding. Both types of T cell carry their own surface receptors that recognise a certain antigen – these make the cell specific for responses involving that antigen. The Ts cells are also specific for different isotypes of Ig, so that they can selectively inhibit, for example, B cells producing IgE antibodies to egg-white protein.
Immunologists believe that the cells in the Peyer’s patch react to microorganisms by inducing both Ts cells and Th cells. These then migrate to the blood, where the balance of effects between them regulates the response.
The Peyer’s patch probably distinguishes food molecules from microorganisms by their smaller size, solubility and lack of ‘stickiness’: microbes have a habit of clinging to cell membranes that could give them away. Exactly how it sorts out the goodies from the baddies is still something of a mystery, however. Food antigens probably stimulate the Peyer’s patch to induce Ts cells, but not Th cells. So the overall effect on the immune response in the blood is inhibitory.
It is possible, however, that the Peyer’s patch stimulates Th cells specific for the non-inflammatory IgA: these would favour IgA antibodies to food molecules in the blood.
The idea that the oral tolerance mechanism breaks down in food-intolerant patients is a popular one, and there is some evidence to support it. The immunoglobulins in the blood specific for food molecules are different from those of healthy people, for example. But the full details of what goes wrong are still far from clear. The levels of SIgA are lower, on average, in those with food intolerance, and this could be a contributory factor. Yet the discovery of individuals with no SIgA at all poses several difficult questions: although such patients mostly have high levels of CICs, and are ill in other ways, few show the symptoms associated with food intolerance.
